Service Communautaire d'Information sur la Recherche et le Développement - CORDIS

Performance and quality management based on MSPC

The aim of the CHEM project was to develop and implement advanced Decision Support Systems (DSS) for process monitoring, data and event analysis, and operation support in industrial processes. The systems are synergistic integration of innovative software tools, which improve the safety, product quality and operation reliability as well as reduce the economic losses due to faulty states, mainly in refining, chemical and petrochemical processes.

The CHEM applications consist of integrated sets of software toolboxes that provide robust detection and diagnosis of process problems in real-time. The systems assist operators in assessing process status and responding to abnormal events. The project provides a flexible architecture and a methodology in order to facilitate the development of such applications on many processes.

Toolbox "Performance and quality management based on MSPC" is a comprehensive implementation of the theory of Multivariate Statistical Process Control (MSPC). MSPC extends standard SPC by taking into account the simultaneous variation of many process variables. The process variables are mapped to a smaller number of orthogonal latent variables, that capture the essential variability of the original data set in a more compact and more easily comprehendible way (for operators and process engineers).

The Toolkit contains all standard functions in MSPC theory, such as PCA (Principal Component Analysis), PLS (Projection to Latent Structures), etc. Generation of MSPC models from raw data, as well as on-line calculation of relevant statistical metrics are supported. APIs are defined for full access of all MSPC functions from G2, hence allowing further processing and reasoning on statistical results. The MSPC modules are embedded in a framework for on-line analysis that emulates sophisticated expert reasoning on the results of the MSPC calculations. The aim is to enable early detection and prevention of performance and quality problems.

"Performance and quality management based on MSPC” is on line at Redcar Blast Furnace to predict aerodynamic instability using a PCA model. The PC's are used to generate a stability index used for both long-term analysis, and to give a minute by minute indication of process stability.

Automatic contribution analysis indicates which parameters have changed significantly when the bi-variate or PC trend is outside its warning limits.

The calculated values are fed into toolbox "Qualitative representation of process trends" and are being assessed for "Qualitative situation assessment". Further results from "Qualitative representation of process trends" are being assessed.

This toolbox is also used to run PLS prediction models for hot metal quality at Redcar Blast Furnace, and for liquor quality from a distillation column at Dawes Lane Coke Ovens, Scunthorpe.

Reported by

Computas AS
Vollsveien 9
1327 Lysaker
See on map